Regional multifocus image fusion using sparse representation

Abstract

Due to the nature of involved optics, the depth of field in imaging systems is usually constricted in the field of view. As a result, we get the image with only parts of the scene in focus. To extend the depth of field, fusing the images at different focus levels is a promising approach. This paper proposes a novel multifocus image fusion approach based on clarity enhanced image segmentation and regional sparse representation. On the one hand, using clarity enhanced image that contains both intensity and clarity information, the proposed method decreases the risk of partitioning the in-focus and out-of-focus pixels in the same region. On the other hand, due to the regional selection of sparse coefficients, the proposed method strengthens its robustness to the distortions and misplacement usually resulting from pixel based coefficients selection. In short, the proposed method combines the merits of regional image fusion and sparse representation based image fusion. The experimental results demonstrate that the proposed method outperforms six recently proposed multifocus image fusion methods.

Multi-Focus Image Fusion using Self-Similarity and Depth Information in Nonsubsampled Shearlet Transform Domain

“…The spatial domain algorithms mainly include principal component analysis [4,5], guided filtering based method [6] and so on. The transform domain algorithms are mainly based on multiresolution geometric analysis (MGA) tool domain, such as image fusion algorithm based on wavelet [7,8], ripplet [9], contourlet [10][11][12], shearlet [13,14], surfacelet [15], trained dictionaries [16,17].…”

Technique for infrared and visible image fusion based on non-subsampled shearlet transform and spiking cortical model

“…Despite a variety of FDA methods, the fundamental distinction among them mainly lies in which category of the frequency domain transform selected. The common transforms include wavelet transform [24,25], lifting stationary wavelet transform [17], quaternion curvelet transform [26], contourlet transform [27], sparse representation [28], shearlet transform (ST) [29], and recently popularized NSCT [30,31]. Although ST and NSCT have much better fusion performance compared with past categories of frequency domain transforms, their inherent drawbacks still cannot be ignored.…”

“…The above seven methods are still implemented as the Refs. [13,15,17,20,22,28,29]. If you want to obtain more details, please see the related contents.…”

Infrared and visible image fusion with spectral graph wavelet transform

“…These classical MSTs have been widely used to achieve infrared-visible image fusion. The main advantages of these methods include offering information on sharp contrast variations to which the human visual cortex is sensitive [15] and preserving the details of different source images well. Thus, MST-based methods can obtain high-quality fused images.…”